Our major aims in Project I are to develop our theoretical perspective on the "practical phonology"[unreadable]that is, on phonological structures of public language use~and to test whether these are the phonological structures accessed by skilled readers of a variety of writing systems. A central requirement of phonological language use is that communicators typically achieve "parity"[unreadable]a relation of sufficient equivalence between phonological messages sent and received. Because language is an evolved system, parity achievement will have been selected for. A parity fostering system is one in which components of the language that serve to make a speaker's message public, that is, phonological forms, are themselves public things. In addition, they are the elements of the message throughout all phases of a communicative exchange. We propose that gestures (linguistically significant actions of the vocal tract) are those phonological primitives. Our research is designed to test and further develop these ideas. We propose to develop our theory of phonological competence, Articulatory Phonology, by showing that it can insightfully address the kinds of phonological systematicities that serve as test cases for phonological theories. Next we show that the primitives of Articulatory Phonology, gestures, are also units of encoding in utterance planning and production. Third we attempt to show that gestures are the primitives of perceived phonological structure. Our research on reading in Project I will extend our exploration of articulatory phonology to the domain of reading. Our proposed reading research develops a critical methodology (masked priming) for tracking the skilled reader's access to phonology from print. Then that methodology among others is used to ask whether the phonology accessed from print crosslinguistically is phonetically-grounded, and to use crossmodal (speech to print) priming to ask whether the phonology accessed from print is that accessed from speech.